Analysing hospital variation in health outcome at the level of EQ-5D dimensions

摘要: The English Department of Health has introduced routine collection patient-reported health outcome data for selected surgical procedures (hip and knee replacement, hernia repair, varicose vein surgery) to facilitate patient choice increase provider accountability. EQ-5D been chosen as the preferred generic instrument current risk-adjustment methodology is based on index score measure variation across hospital providers. There are two potential problems with this. First, using a population value set generate may not be appropriate purposes performance assessment because it introduces an exogenous source assumes identical preferences dimensions among patients. Second, multimodal distribution creates statistical that yet resolved. Analysing each dimension (mobility, self care, usual activities, pain/discomfort, anxiety/depression) seems therefore more promising. For hip replacement surgery, we explore a) impact treatment b) extent which varies providers c) correlated within We combine information pre- post-operative outcomes Hospital Episode Statistics financial year 2009/10. overall sample consists 25k patients complete responses. employ multilevel ordered probit models recognise hierarchical nature (measurement points nested in patients, themselves providers) response distributions. modelled random coefficient at hospital-level. obtain provider-specific Empirical Bayes (EB) estimates this coefficient. estimate separate five analyse correlations EB dimensions. Our analysis suggests indeed associated improvements variability generally pronounced mobility, activity pain/discomfort than others. pairwise correlation between substantial, suggesting certain better improving multiple others multivariate should further investigated.